This data set consists of the taxonomically harmonized and temporally standardized fossil pollen data from 3680 records. 1122 records are located in North America, 1446 records in Europe, 687 records in Asia, 185 records in South America, 159 in Africa and 81 in the Indo-Pacific region. We expanded the previous version of the LegacyPollen 1.0 data set (Herzschuh et al., 2022; https://doi.org/10.5194/essd-14-3213-2022) with records from the Neotoma Paleoecology Database (https://www.neotomadb.org/; last access: August 31, 2022), ACER 1.0 database (Sánchez Goñi et al., 2017; https://doi.org/10.1594/PANGAEA.870867), Chinese fossil pollen dataset (Zhou et al., 2023, https://www.plant-ecology.com/CN/Y2023/V47/I10/1453 [in Chinese]; Cao et al., 2022, https://doi.org/10.1111/gcb.16274), and our own collection for the Asian sector. Taxonomic harmonization (i.e., woody taxa and major herbaceous taxa have been harmonized to genus level and other herbaceous taxa to family level) and temporal standardization (i.e., re-estimation of age-depth models) follow the previously established frameworks LegacyPollen 1.0 (Herzschuh et al., 2022; https://doi.org/10.5194/essd-14-3213-2022) and LegacyAge 1.0 (Li et al., 2022; https://doi.org/10.5194/essd-14-1331-2022), respectively. In compiling the dataset, we also followed the practices recommended by Flantua et al. (2023; https://doi.org/10.1111/geb.13693) for large-scale paleoecological data synthesis, such as how to select data sources and filter the dataset.Compared to the LegacyPollen 1.0 dataset, we now include the Neotoma DOI (if Neotoma source) in the overview table of site metadata to eliminate the broken chain of static LegacyPollen 2.0 dataset with living (such as updating discovered metadata errors and chronologies) Neotoma and associated risk of data staleness. Furthermore, we also added the PANGAEA Event (PANGAEA dataset identifier) for each new record to ensure that our dataset meets PANGAEA's high standards for quality, usability, and compliance.